CuanReply

CuanReply is an AI-powered business assistant for resellers that turns stagnant chat data into actionable sales strategies using Amazon Bedrock and DynamoDB.

Comment

Inspiration

Many resellers struggle with high volumes of customer inquiries, leading to burnout and missed sales opportunities. We saw that chat data was being wasted rather than used as a strategic asset. This inspired us to build CuanReply, a tool that turns daily conversations into actionable business intelligence.

What it does

CuanReply serves as an AI-powered assistant that helps resellers manage inquiries efficiently. It provides a Multi-Perspective Strategy feature, giving users various strategic response options to convert leads into sales. Every interaction is securely stored to build a persistent business history.

How we built it

The project is built on a modern full-stack architecture:

  • [cite_start]Frontend: Crafted with HTML5, CSS3, and Vanilla JavaScript for a fast, responsive UI[cite: 13, 14, 15].
  • [cite_start]Intelligence: Integrated Amazon Bedrock using the Nova Lite model to generate intelligent, low-latency responses[cite: 16, 17, 18].
  • [cite_start]Backend: Developed a high-performance server using Node.js and Express.js[cite: 19, 20].
  • [cite_start]Database: Utilized Amazon DynamoDB for scalable and reliable cloud data persistence[cite: 21, 22].

Challenges we ran into

One of the biggest hurdles was managing AWS permissions and ensuring seamless data flow between the intelligence model and the database. We also focused heavily on optimizing the UI to handle long chat histories while maintaining a "fresh" and simple user experience.

Accomplishments that we're proud of

We successfully integrated a state-of-the-art LLM (Nova Lite) into a functional web application that actually solves a real-world problem for MSMEs. Seeing the AI provide diverse strategic perspectives was a huge milestone for us.

What we learned

We deepened our understanding of the AWS ecosystem, particularly how to leverage Bedrock for generative AI and DynamoDB for NoSQL data management. We also learned the importance of "Design Thinking" in creating tools that are not just high-tech, but also user-friendly.

What's next for CuanReply

Our roadmap includes implementing real-time sentiment analysis, expanding to platforms like WhatsApp and Telegram, and developing an advanced analytics dashboard for deeper business insights.

Share this project:

Updates